Open Access Open Access  Restricted Access Subscription Access

Pedestrian Detection

Mrs.Tejasvini Shinde, Tanuja Patil, Lubdha Rajmane, Suzan Shaikh, Siddhi Thombare, Saima Waikar

Abstract


Pedestrian detection technology is a critical component of modern intelligent transportation systems, acting as an advanced driver assistance system (ADAS) designed to identify people in a vehicle's path. Intelligent surveillance, sophisticated driver assistance systems, and driverless cars all depend on pedestrian detection. The pedestrian identification system presented in this research is based on deep learning and is intended to operate in complicated urban contexts with high accuracy and real-time performance. The suggested method efficiently detects pedestrians under various illumination conditions, occlusions, and scale variations by using convolutional neural networks with multi-scale feature extraction. Techniques for data augmentation and transfer learning are used to improve the generalization and resilience of the model. When tested on common benchmark datasets, the system shows better detection accuracy and fewer false positives than conventional techniques. The suggested framework offers a dependable and effective solution for practical pedestrian detection applications, according to experimental results.

 


Full Text:

PDF

References


Dalal, N., “Histograms of Oriented Gradients for Human Detection”, Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR), vol. 1, pp. 886–893, 2005.

Redmon, J., “YOLOv3: An Incremental Improvement”, arXiv preprint arXiv:1804.02767, 2018

OpenCV, “Object Detection (Haar-cascade)”, OpenCV Documentation, available at: https://docs.opencv.org/, accessed 10 August 2025.

Viola, P., “Rapid Object Detection using a Boosted Cascade of Simple Features”, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR), vol. 1, pp. I-511, 2001.

Zhang, L., Lin, L., Liang, X., & He, K.,“From Handcrafted to Deep Features for Pedestrian Detection: A Survey,” IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016.

Rahman, M., Islam, I., Chowdhury, M., Dunning, J., Wood, S., Rice, D., & Rizenbergs, C.,“Real-Time Pedestrian Detection with Edge Computing & YOLOv3,” IEEE Transactions on Intelligent Transportation Systems, 2020.

García, F., Pérez, J., Muñoz, J., & López, A.,“Deep Learning-Based Pedestrian Detection in Autonomous Vehicles: Issues & Challenges,” IEEE Transactions on Intelligent Vehicles, 2021.

Terven, J., & Córdova-Esparza, D.,“From YOLOv1 to YOLOv8 & YOLO-NAS: A Comprehensive Review,” ACM Computing Surveys, 2023.

Kyrkou, C.,“YOLOpeds: Efficient Real-Time Single-Shot Pedestrian Detection,” IEEE Embedded Systems Letters, 2020.

Wang, H., Nie, Y., Li, Y., Liu, H., Liu, M., & Cheng, W., “Research, Applications and Prospects of Event-Based Pedestrian Detection: A Survey,” Sensors, 2022.


Refbacks

  • There are currently no refbacks.